Pivotal Response Treatment Increases Neural Processing Efficiency of Faces in
Children with Autism Spectrum Disorder
Zachary Williams, Max Rolison, Katherine Stavropoulos, Jennifer Foss-Feig, Sabrina Malak, Adam Naples, Kevin Pelphrey, Pamela Ventola & James McPartland
Yale Child Study Center, New Haven, CT
Statistical Analysis:
•  Peak amplitude and latency of P1/N170 and average amplitude
of P3 were analyzed using repeated measures ANOVA
•  2 within-subjects factors:
•  Treatment (Pre/Post)
•  Emotion (Fear/Neutral)
•  Left-hemisphere and right-hemisphere N170 components were
analyzed individually (LN170 and RN170, respectively)
•  Paired samples t-tests for Waitlist Control vs. Pre and Follow-up
vs. Post conditions
5.6 (0.91)
DAS-II 112.3 (7.8) 112.3 (11.4)
ADOS 17.3 (3.2)
17.9 (6.5)
12.4 (5.4)
Table 1: Group Statistics: Subject Age, ADOS
severity score and IQ (DAS-II GCA) at waitlist (N = 3),
pre-treatment (N = 7), and post-treatment (N = 7)
16
14
Left - Neutral
N170
12
16
Pre
Post
Follow-up
14
10
10
8
8
6
4
2
0
0
-2-100
Latency (ms)
-4
100
200
300
400
-6
16
14
N170
12
EEG Data Acquisition and Collection:
•  Recorded at 500 Hz
•  128-channel Hydrocel Geodesic Sensor net
Figure 1: Experimental Paradigm. Static neutral or fearful
faces were presented for 500 ms before they dynamically
shifted into a different facial expression. ERP data was
segmented to the onset of the static face.
Figure 2: P1, N170, and P3 electrode
recording sites. Data were averaged
across all electrodes in each cluster (P1,
Left N170, Right N170, P3)
Pre
Post
Follow-up
N170
6
4
2
0
0
-2-100
Latency (ms)
-4
100
200
300
400
200
300
400
Waitlist Control (WLC)
•  Subset of 3 children
•  P1 latency showed marginally significant increase
over waitlist control period (p = .087)
•  No significant change in RN170 latency across
waitlist period for either emotion (p’s > .123)
•  LN170 peak amplitude was increased over the
waitlist control period when children viewed fearful
(p = .017) but not neutral (p = .322) faces
•  No significant changes in P1 amplitude, LN170
latency, RN170 amplitude, or P300 average
amplitude over this period (all p’s > .213)
10
5
0
-100
-5
0
100
200
300
400
Latency (ms)
-10
2
0
0
-2-100
Latency (ms)
-4
• 
• 
• 
• 
• 
100
200
300
400
Pre
Post
Follow-up
Right - Fear
N170
12
8
6
4
2
0
0
-2-100
Latency (ms)
-4
100
200
300
400
-6
Figure 3: Grand average waveform depicting the Left and Right N170 across all participants (N= 7,
N=5 for follow-up) for fear and neutral faces, at pre-treatment, post-treatment, and 16-week follow-up
Follow-up (N = 5)
All Trials
Neutral
Fear
Figure 5: RN170 latency in the subset of children
who completed a follow-up assessment (N = 5)
WLC (N = 3)
Pre (N = 3)
All Trials
Neutral
Fear
Figure 6: LN170 amplitude in waitlist controls (N = 3)
4
10
8
-6
Right - Neutral
14
10
Latency (ms)
100
Follow-Up
•  Subset of 5 children
•  Reduction LN170 peak amplitude did not change
significantly 16 weeks post-treatment (p = .454)
•  RN170 latency showed a marginally significant
increase from post-treatment to follow-up (p = .071)
•  Latency increase was not significant for either
fear (p = .112) or neutral (p = .555) faces alone
•  P1 latency also did not differ between posttreatment and follow-up (p = .104)
6
16
Pre
Post
Follow-up
-5
0
Pre
Post
Follow-up
P1
Figure 4: Grand average waveforms depicting the P1 across participants for fear and neutral faces, pretreatment, post-treatment and at follow-up
Post (N = 5)
-6
Left - Fear
0
-100
-10
12
Amplitude (µV)
5.7 (0.62)
Post
Amplitude (µV)
•  Artifact-free data was successfully collected
from 5 out of the 7 children at a follow-up
evaluation 16 weeks after the end of
treatment
•  Received PRT for 16 weeks
•  8 hours per week (6 hours with the child and
2 hours of parent guidance)
Experimental Paradigm:
•  Participants viewed computer-generated faces
showing neutral and fearful affect
•  Attentional task (button press upon seeing
bouncing ball) interspersed between trials
•  EEG recorded at 4 time points:
•  Waitlist Control
•  Pre-Treatment
•  Post-Treatment
•  Follow-Up (16 weeks after treatment end)
Pre
Amplitude (µV)
•  Subset of 3 served as waitlist control
(WLC) prior to enrollment in treatment
Age
WLC
Amplitude (µV)
Participants:
•  7 children 4-6 year of age with ASD receiving
PRT
5
Fear
15
10
Post-Treatment
P1
•  No significant effects on P1 amplitude (all F's < 1.16, p's > .324)
•  Significant treatment effect on P1 latency [F(1, 6) = 7.269, p = .036, η2p = .548], qualified by a
significant treatment x emotion interaction [F(1, 6) = 6.422, p = .044, η2p = .517]
•  Reduction in P1 latency with treatment was significant for neutral faces (p = .021) but not fearful
faces (p = .495)
N170
•  A main effect of treatment [F(1,6) = 6.34, p = .045, η2p = .514] indicated a change in face processing
efficiency following PRT treatment, indexed by right-hemisphere N170 latency
•  Reduction in RN170 latency following treatment only trended toward significance for either
neutral (p = .065) or fearful (p = .092) faces alone
•  No similar effect found in the left-hemisphere N170 latency (p = .160)
•  Left-hemisphere N170 peak amplitude was significantly reduced from pre- to post-treatment [F(1, 6)
= 14.217, p = .009, η2p = .703], reflecting an attenuated response in the hemisphere not typically
associated with face processing
•  Significant for both fearful (p = .039) and neutral (p = .008) faces alone
•  There was no effect of treatment on RN170 peak amplitude (p = .311)
•  No significant effects of emotion or emotion * treatment interactions on N170 amplitude or latency (all
F's < .88, p's > .382)
P3
•  No significant effects of treatment, emotion or emotion * treatment interactions on P3 average
amplitude (all F's < .62, p's > .459)
We predicted that children would demonstrate improved social processing efficiency, as reflected in
decreased latency of the N170 component, following a 4-month course of PRT. We also expect to see
an increased level of processing of the stimuli, as indexed by an increase in the average P3
amplitude. These effects were expected to persist beyond the end of treatment. No changes in P1
amplitude or latency were expected.
20
Latency (ms)
The current study utilized event-related potentials (ERPs) to examine the temporal dynamics of social
perception following a course of the PRT intervention
•  P100 (P1)
•  Positive-going component recorded over occipital scalp approximately 100ms after viewing an
image (Boutsen et al., 2006)
•  Believed to reflect processing of the low-level features of stimuli (Rossion & Caharel, 2011)
•  Amplitude modulated by attention (Herrmann & Knight, 2001)
•  N170
•  Component with negative peak around 170 ms in the occipitotemporal regions
•  Peak is generally later in young children (Taylor et al., 2001)
•  Preliminary perceptual encoding of faces underlying categorization
•  When typically developing individuals view faces, the N170 component is of larger magnitude
in the right hemisphere
•  P300 (P3)
•  Large-amplitude positive component measured on parietal midline approximately 300ms after
stimulus onset
•  Thought to reflect processing of information when it is incorporated into memory
representations of the stimulus and context (Polich & Herbst, 2000)
Pre
Post
Follow-up
P1
Amplitude (µV)
ERP Analysis:
•  Data segmented to static face presentation within larger
dynamic face paradigm
•  Component statistics derived from average values across all
electrodes in a particular cluster and time window
•  P1 window: 80-150 ms
•  N170 window: 164-290 ms
•  P3 window: 250-390 ms
Neutral
15
Amplitude (µV)
Pivotal Response Treatment (PRT) is an evidence-based, naturalistic behavioral intervention derived
from Applied Behavior Analysis. PRT targets "pivotal areas" of a child's development, domains in
which improvement produces collateral gains in other areas not specifically addressed by the
intervention.
•  A 4-month course of PRT results in meaningful improvements in pragmatic language, social
engagement, and adaptive functioning (Ventola et al., 2014)
•  More normative neural responses in a biological motion perception task were observed following
PRT, as indexed with fMRI (Ventola et al., 2015)
•  To date, post-PRT changes in the temporal dynamics of neural activation have yet to be studied
20
P1
N170
P3
Amplitude (µV)
Autism Spectrum Disorder (ASD) is a chronic neurodevelopmental condition characterized by a
cluster of deficits in social communication and the presence of restricted and repetitive behaviors and
interests.
• 
A 16-week course of PRT for young children with ASD resulted in improved efficiency of neural
indicators of social perception (RN170 latency), as well as a less atypical lateralization of the facesensitive N170 response. These findings suggest focal treatment effects on social brain processes
A decrease in the latency of a low-level sensory ERP component (P1) to neutral faces was observed
post-treatment, suggesting that PRT caused a more rapid orienting of processing resources to
neutral face stimuli
No significant effects on P3 amplitude were observed, perhaps due to large inter-subject variability
of the component’s amplitude and shape over the chosen latencies
Follow-up data showed no significant changes in P1 latency, LN170 amplitude, or RN170 latency
between the end of treatment and a 16-week follow-up evaluation, supporting the hypothesis that the
effects of treatment persist beyond the duration of the intervention
•  RN170 latency at follow-up showed a marginally significant increase from post-treatment,
indicating that ongoing treatment may be necessary to maintain this effect
•  Further study is warranted to relate post-treatment brain responses to loss of behavioral gains
from treatment
The preliminary waitlist control results show values trending in the opposite direction of those of the
children receiving the intervention, suggesting that observed changes are not simply a function of
development
•  Several effects in the waitlist and follow-up subgroups showed trends toward significance,
highlighting the need for larger samples to confirm these findings
These findings provide the first evidence of improved neural efficiency resulting from PRT
•  In concert with fMRI results following a 16-week course of PRT, these ERP findings inform
understanding of brain mechanisms underpinning positive response to behavioral treatment
References
Boutsen, L., Humphrey, G. W., Praamstra, P., & Warbrick, T. (2006). Comparing neural correlates of configural processing in faces and objects: An ERP study of the thatcher illusion.
NeuroImage, 32(1), 352–367.
Herrmann, C. S., & Knight, R. T. (2001). Mechanisms of human attention: event-related potentials and oscillations. Neuroscience & Biobehavioral Reviews, 25(6), 465-476.
Polich, J., & Herbst, K. L. (2000). P300 as a clinical assay: rationale, evaluation, and findings. International Journal of Psychophysiology, 38(1), 3-19.
Rossion, B., & Caharel, S. (2011). ERP evidence for the speed of face categorization in the human brain: Disentangling the contribution of low-level visual cues from face perception.
Vision Research, 51(12), 1297–1311.
Taylor, M. J., Edmonds, G. E., McCarthy, G., & Allison, T. (2001). Eyes first! Eye processing develops before face processing in children. Neuroreport, 12(8), 1671-1676.
Ventola, P., Friedman, H. E., Anderson, L. C., Wolf, J. M., Oosting, D., Foss-Feig, J., . . . Pelphrey, K. A. (2014). Improvements in social and adaptive functioning following short-duration
PRT program: a clinical replication. J Autism Dev Disord, 44(11), 2862-2870.
Ventola, P., Yang, D. Y., Friedman, H. E., Oosting, D., Wolf, J., Sukhodolsky, D. G., & Pelphrey, K. A. (2015). Heterogeneity of neural mechanisms of response to pivotal response
treatment. Brain imaging and behavior, 9(1), 74-88.
Acknowledgments
NIMH R01 MH100173 (McPartland); NIMH K23 MH086785 (McPartland); NIMH R21 MH091309 (McPartland); NIMH R01 MH100028 (Pelphrey); CTSA Grant Number UL1 RR024139
(McPartland); Allied World (Ventola & Pelphrey); Harris Professorship at the Yale Child Study Center (Pelphrey); Deitz Family (Ventola); Women’s Health Research at Yale (Ventola);
Simons Foundation (Ventola); Esme Usdan and Family (Ventola); Autism Science Foundation Summer Undergraduate Research Grant (Williams); INSAR Travel Award (Williams)